Yes, I had bad sector on md3 (raid10) array. Disk was replaced. But it
looks like it is unrelated to the reason why my box is hanging. It
started after kernel update from 3.10 to 3.14.2 and vbox to 4.3.10.
Now i updated kernel to 3.14.3, going 4h from update and it looks like
my problem is gone.
Regards,
Łukasz
On 05/13/14 15:39, Nowaker wrote:
I recently updated kernel to 3.14.2 and now I see i/o subsystem hang in
few minutes after I start VirtualBox 4.3.10 machines (with Windows7).
messages.log below. It looks like every process that tries to access
/dev/md3 (even kworker and md3_resync) hangs forever.
Anyone has similar problems?
I also use soft raid1 so I just ran W7 on my 3.14.3 and it just works.
You are probably hitting some bad sector. VM disk is on /dev/md3, isn't
it? You may want to `cat` read the VM disk file, and dd `read` the whole
drives and the md drive to see if they are readable. You may want to
inspect /proc/mdstat, /sys/block/**/bad_blocks and smartctl reports.
You may be facing some bug in the raid drivers. I faced one too, which
was then fixed. https://bugzilla.kernel.org/show_bug.cgi?id=68181 If it
*is* a bug, the best what you can do is to report the bug, and keep the
the current raid mirror as long as possible so you can tests patches the
kernel guys provide.
